Cervical cancer, human papillomavirus, and vaccination

At the beginning of October one of the trials of vaccines against human papillomavirus (HPV) infection, the primary risk factor for cervical cancer, announced an encouraging result.¹ Large scale, multi-country, multi-site trials of several HPV vaccines are currently under way. The end points comprise incident and persistent HPV infection (during 2-3 years' follow-up) and associated precancerous cytological and histological lesions (cervical intraepithelial (CIN) neoplasia during 2-3 and 4-5 years' follow-up). The World Health Organization is expecting at least one of these vaccines to be licensed for use in 2006.² How promising are the available trial results? What other questions need answering? And is it time to accelerate preparations for programmes to provide HPV vaccination?

The latest data release concerns a trial of a quadrivalent recombinant vaccine that included HPV types 6, 11, 16, and 18.¹ In all, 12 167 women at 90 centres in 13 countries participated in the trial, the FUTURE II study. In this prospective double blind study, women aged 16 to 26 were randomised to receive three doses of either vaccine or placebo over six months.

In a restricted “according to protocol” analysis that included only women who had no protocol violations and who received a full course of vaccine and were free of infection with HPV types 16 and 18 at month 7, the reported efficacy of the vaccine after an average follow-up of 17 months was 100%. In the 5301 vaccinated women there were no observed cases of high grade precancer or non-invasive cancer (CIN (cervical intraepithelial neoplasia) 2 or 3 or adenocarcinoma in situ) related to HPV 16 or 18, but there were 21 cases in the 5258 women given placebo. Protective efficacy was 97% against CIN 2 or 3 or adenocarcinoma in situ in a broader analysis that covered the period from one month after the first dose of vaccine to an average follow-up of two years. This analysis included the women in the restricted analysis group as well as those who might have been infected with HPV 16 or 18 during vaccination or whose management during the trial violated the protocol. In the broader analysis, one case of CIN 2 or 3 or adenocarcinoma in situ was observed in the 5736 vaccinated women compared with 36 in the placebo group.

These data from FUTURE II are the most recent in a series of highly promising trial results of monovalent (HPV 16), bivalent (HPV 16, 18) and quadrivalent (HPV 6, 11, 16, 18) vaccines.^3-5 Each of the vaccine formulations studied has used recombinant non-infectious virus-like particles comprising protein subunits expressed in yeast or baculovirus. Reported side effects in all the vaccinated groups were mostly limited to mild symptoms at the injection site, which did not affect overall compliance. No serious adverse events attributable to the vaccines have occurred.

Of the more than 100 types of HPV that have been described, about 40 can infect the epithelial lining of the anogenital tract. Infection with “low risk” HPV types can result in anogenital warts, benign genital lesions, or persistent subclinical infection. At least 95% of cervical cancer results from infection with the 15 or more HPV types classified as high risk, with HPV 16 and 18 together accounting for around 70% of all cervical cancer.⁶,⁷ Although cervical cancer is a relatively rare consequence of HPV infection and most HPV infections resolve spontaneously, the incidence of HPV infection is so high in sexually active adults that, worldwide, cervical cancer remains the second most common cancer among women, after breast cancer.⁸

A recent pooled analysis showed the age standardised prevalence of all types of HPV infection to vary 20-fold among different regions of the world. The prevalence of high risk types of the virus was 18% in sub-Saharan Africa, 5% in Asia, 10% in South America, and 4% in Europe.⁹ As a probable consequence, the incidence of cervical cancer is greatest in poorer countries in Africa and Latin America, where it is generally the most common cause of death from cancer in women.⁸

An analysis of trends in mortality from cervical cancer before and after cervical screening was introduced in England suggests that up to 4500 lives are saved each year as a result of screening.¹⁰ Nevertheless, 2424 new cases of invasive cervical cancer occurred in 2000, and 927 deaths from invasive cervical cancer were registered in 2002.¹¹ A vaccination programme for HPV 16 and 18 could further reduce the incidence of and mortality from cervical cancer in England and other countries, and it may considerably reduce the costs of maintaining screening programmes. A vaccine that also protected against anogenital warts would decrease the costs of treatment and reduce the workload of sexual health services.

Questions before starting an HPV vaccination programme

• What proportions of cervical cancer and other HPV related disease in a region or country are attributable to the HPV types targeted by the available vaccines?

• What fraction of cervical cancer overall will be prevented by a vaccine against HPV 16 and 18?

• Will immunity induced by vaccines alter the distribution of other, non-vaccine HPV types?

• Will a vaccination programme against a sexually transmitted infection prove acceptable to adolescents who are not yet sexually active and their parents?

• Should teenage boys be vaccinated as well as teenage girls?

• Will booster vaccinations be necessary, and if so, when?

• How will a vaccination programme affect current programmes for cervical cancer screening, and how and when should screening change in response?

• What benefits might vaccination confer on adults who are already infected with HPV?

• Should older sexually active adults be included as part of a catch-up campaign at the outset of a vaccination programme?

• Should any catch-up campaign be aimed at specific subgroups of the population?

• What will be the cost effectiveness of various strategies for vaccination programmes?

Important questions need answering before a programme of HPV vaccination can be introduced (box). Furthermore, because the incubation period between HPV infection and development of invasive cervical cancer is long, prevention of cancer by a vaccination programme will not be obvious for 10 to 20 years. Earlier, the rates of abnormal cervical smears showing early stages of CIN will diminish, and the need for follow-up after screening will be reduced.

Efforts at prevention have had a limited effect so far on the occurrence of sexually transmitted infections.¹² Current trends in sexual behaviour, with 44% of women and 52% of men aged 16 to 24 in Britain reporting three or more sexual partners in the past five years,¹³ are probably increasing the incidence of HPV infection. The need for a programme of HPV vaccination can only intensify, and every country should be planning actively to implement and budget for early vaccination. The vaccines under evaluation already have the potential to prevent a substantial proportion of cervical cancer cases, and a second generation of HPV vaccines containing additional high risk types and combined with other vaccines is being developed. Should these first and second generation HPV vaccines prove as effective as they promise, we hope that the new International Finance Facility for Immunisation¹⁴ will be used to accelerate their delivery in the poorer parts of the world, where the benefits will be greatest.